Interbreeding between early human groups, such as Neanderthals and Denisovans, and anatomically modern humans (Homo sapiens) occurred during the Middle Paleolithic and early Upper Paleolithic periods. Studies of DNA have shown that all modern human populations outside Africa today carry about 1–4% Neanderthal DNA. This DNA was inherited after modern humans left Africa and mixed with Neanderthals. Denisovan DNA is most common in Oceania, where modern humans have about 4–6% Denisovan DNA in their genomes. People in Eurasia and the Americas have smaller amounts of Denisovan DNA.
In Africa, later migrations from Eurasia during the Neolithic period brought Neanderthal DNA to North African populations. Earlier, it was believed that modern humans in sub-Saharan Africa had no Neanderthal DNA. However, recent research has found very small amounts of Neanderthal DNA in these populations as well.
The mixing of archaic human DNA with modern human DNA has affected human biology through both helpful and harmful genetic changes. Some genetic traits from archaic humans helped modern humans survive in new environments, such as genes related to the immune system, skin and hair features, and adaptations to high altitudes. However, certain parts of the modern human genome, especially on the X chromosome and in genes active in testes, show no archaic DNA. This suggests that harmful genetic traits from these mixtures were removed over time through natural selection, likely because they reduced fertility in male hybrids.
These discoveries have changed scientific understanding of human history. Previously, the idea that modern humans replaced archaic humans without mixing was widely accepted. However, DNA evidence since 2010 has shown that while most modern human ancestry comes from Africa, populations today have been shaped by ancient interbreeding with archaic humans in Africa, Eurasia, and other regions. This has shifted the scientific view from a strict "Out of Africa" replacement model to one that includes hybridization events.
Neanderthals
On May 7, 2010, scientists published a draft sequence of the Neanderthal genome. This was based on the genome sequencing of three Vindija Neanderthals. The study showed that Neanderthals shared more genetic traits with Eurasian populations than with sub-Saharan African populations. The researchers believed that this similarity was best explained by gene flow from Neanderthals to modern humans after humans left Africa. They estimated that 1–4% of the Eurasian genome came from Neanderthals. Other studies later gave slightly different estimates, such as 1–6% and 1.5–2.1%. In 2017, one study revised the estimate to 1.8–2.6% for non-Africans outside Oceania.
A later study by Chen et al. (2020) found that Africans also have Neanderthal genetic material. This admixture was estimated to be 0.3% of their genome, or 17 megabases. The researchers suggested that Africans gained this genetic material from a back-migration of modern humans who had Neanderthal ancestry. This migration is thought to have occurred about 20,000 years ago. However, some scientists, like David Reich, believe the genetic signal from Neanderthal admixture in Africa is very weak and may not be as significant as previously thought.
Studies have shown that 50% of the Neanderthal genome is present in people from India, and 41% in Icelanders. Earlier research estimated that about 20% of the Neanderthal genome was found in modern Eurasians, though some studies suggested a third. A 2023 study found evidence of gene flow from modern humans to Neanderthals around 250,000 years ago. It estimated that about 6% of the Altai Neanderthal genome was inherited from modern humans.
East Asians were found to have a higher level of Neanderthal admixture than Europeans, with about 20% more genetic material from Neanderthals. This could be due to additional admixture events in the ancestors of East Asians after they separated from Europeans, or because of reduced effectiveness of natural selection against Neanderthal genes in East Asian populations. Studies suggest that a single pulse of Neanderthal admixture after humans left Africa is the most likely explanation for the higher levels in East Asians. However, differences in Neanderthal ancestry rates are smaller than previously thought, and other factors may also contribute.
Genomic analysis shows that Neanderthal genetic material is more common in non-African populations than in sub-Saharan African populations. North African groups share a similar level of Neanderthal genetic traits with non-African populations as they do with Neanderthals. In contrast, sub-Saharan African populations generally do not show evidence of Neanderthal admixture. The amount of Neanderthal genetic material in North African populations varies depending on their ancestry, with higher levels found in groups with more North African ancestry. Studies suggest that the Neanderthal genetic signal in Africa is not due to recent gene flow from Europe or the Near East but may be linked to ancient North African populations.
A small but significant variation in Neanderthal admixture rates was observed among European populations, but not among East Asian populations. Prüfer et al. (2017) found that East Asians carry more Neanderthal DNA (2.3–2.6%) than Western Eurasians (1.8–2.4%).
Chen et al. (2020) revised earlier estimates, finding that East Asians have 8% more Neanderthal ancestry than Europeans, not 20% as previously reported. This change was due to a flawed assumption that Africans had no Neanderthal admixture, which led to an underestimation of Neanderthal ancestry in non-African populations. The study suggested that a single pulse of Neanderthal admixture after humans left Africa best explains the higher levels in East Asians.
Genomic analysis also shows that the Neanderthal genetic material in non-African modern humans is more closely related to the Mezmaiskaya Neanderthal (from the North Caucasus) than to the Altai Neanderthal (from Siberia) or the Vindija Neanderthals (from Croatia). High-coverage sequencing of a Vindija Neanderthal genome showed that the Vindija and Mezmaiskaya Neanderthals shared similar levels of genetic traits with modern humans. This suggests that most of the Neanderthal admixture into modern humans came from populations that diverged from the Vindija and Mezmaiskaya lineages about 80,000–100,000 years ago.
Analysis of chromosome 21 in the Altai, El Sidrón (Spain), and Vindija Neanderthals showed that only the El Sidrón and Vindija Neanderthals had significant gene flow into modern humans (0.3–2.6%). This suggests that the El Sidrón and Vindija Neanderthals are more closely related to the Neanderthals that interbred with modern humans about 47,000–65,000 years ago than the Altai Neanderthal.
Denisovans
Studies show that Melanesians, such as people from Papua New Guinea and Bougainville Island, share more gene versions with Denisovans compared to other groups like Eurasians and Africans. Scientists estimate that 4% to 6% of the genes in Melanesians come from Denisovans. However, no Eurasians or Africans have Denisovan genes in their DNA. Researchers have found that Denisovans passed some genes to Melanesians but not to East Asians. This suggests that early ancestors of Melanesians interacted with Denisovans, but this interaction likely happened outside of southern Siberia, where Denisovan remains have been found. Aboriginal Australians also share more Denisovan gene versions than Eurasians and Africans, which supports the idea that Denisovans and Melanesians had more contact.
Reich and others (2011) found that Oceanian populations, such as Aboriginal Australians, Near Oceanians, and Polynesians, have the highest levels of Denisovan genes. Some eastern Southeast Asian populations also show Denisovan genes, but mainland East Asian and African populations do not. This suggests that Denisovans may have mixed with early humans in Southeast Asia, not in mainland Eurasia. The lack of Denisovan genes in mainland Asia and the high levels in Oceania indicate that Denisovans and early humans interbred east of the Wallace Line, which divides Southeast Asia.
Skoglund and Jakobsson (2011) noted that Oceanians and Southeast Asians have the most Denisovan genes compared to other groups. They also found possible traces of Denisovan genes in East Asians but not in Native Americans. However, Prüfer and others (2013) found that mainland Asians and Native Americans may have a very small amount of Denisovan genes, about 0.2%, which is much less than in Oceanians. Wall and others (2013) found no evidence of Denisovan genes in East Asians.
Research suggests that Denisovans shared genes with the ancestors of Aboriginal Filipinos, Aboriginal Australians, and New Guineans. New Guineans and Australians have similar levels of Denisovan genes, which means interbreeding likely happened before their ancestors moved to Sahul (Pleistocene New Guinea and Australia), at least 44,000 years ago. Some populations, like those in Nusa Tenggara, Moluccas, Polynesia, and Fiji, have Denisovan genes in amounts similar to their Near Oceanian ancestry. However, the Philippine Mamanwa and Manobo groups have higher Denisovan ancestry than Near Oceanian ancestry. Reich and others (2011) proposed a model where early humans migrated east, some interbred with Denisovans, and later groups diverged.
Browning and others (2018) found evidence that Denisovans mixed with humans in at least two different events. East Asians, such as Japanese and Han Chinese, show signs of mixing with two different Denisovan groups. South Asians, like Telugu and Punjabi people, and Oceanians, like Papuans, show signs of mixing with one Denisovan group.
Sankararaman and others (2016) estimated that Denisovans mixed with humans 44,000 to 54,000 years ago. Oceanians had the most Denisovan genes compared to other groups with Denisovan ancestry, such as those in America, Central Asia, East Asia, and South Asia. South Asians also had more Denisovan genes than other non-Oceanian groups, though less than Oceanians. Researchers suggest this could mean one mixing event that spread differently or multiple events.
A 2021 study found that Denisovans mixed with Philippine Negritos, such as the Ayta Magbukon, who have the highest Denisovan ancestry in the world, about 30% to 40% more than Australians and Papuans. This suggests Denisovans in the Philippines mixed with humans after they arrived.
Eurasians have some but much less Denisovan-related genes than Oceanians. This is because Denisovans are related to Neanderthals, who contributed genes to Eurasians, not because Denisovans mixed directly with Eurasian ancestors.
A 40,000-year-old human skeleton from the Tianyuan cave in China showed Neanderthal genes but no Denisovan genes. This suggests Denisovan genes were rare in mainland Asia.
Some parts of the human genome lack Denisovan genes, possibly because male hybrids between Denisovans and humans were less likely to survive. This is supported by fewer Denisovan genes on the X chromosome and in testes-related genes.
Studies of immune system genes (HLA) suggest that HLA-B73 may have come from Denisovans in western Asia. Though not found in the Denisovan genome, HLA-B73 is linked to Denisovan-derived HLA-C15:05. Analysis suggests HLA-B73 is an ancient allele.
Denisovans had HLA-A (A02 and A11) and HLA-C (C15 and C12:02) genes that are common in modern humans. One HLA-B gene in Denisovans is rare, and another is missing in modern humans. Scientists believe these genes were passed from Denisovans to humans, as it is unlikely they would have developed independently due to high mutation rates.
Tibetan people have a helpful gene variant from Denisovans, called EGLN1 and EPAS1, which helps them live at high altitudes.
Archaic African hominins
Fossils in Sub-Saharan Africa decay quickly, making it difficult to compare modern human DNA with ancient DNA from early human ancestors in the same region.
Ancient DNA from a person in Ethiopia (about 4,500 years old) and from other areas in Africa (between 1,300 and 8,100 years old) shows that some West African populations have small amounts of DNA that do not match the genetic patterns of early hunter-gatherers in Eastern or Southern Africa. This suggests that these West African groups, such as the Yoruba from Nigeria and the Mende from Sierra Leone, inherited this DNA long before farming spread and likely before the start of the Holocene (about 11,600 years ago). This ancient DNA must have come from a group that separated from the ancestors of the San people (a group in Southern Africa) at least 200,000 to 300,000 years ago.
Other studies support the idea that some modern African populations have DNA from an ancient group that existed before the San, Pygmies (from Central Africa), and early African hunter-gatherers. This evidence comes from research on long DNA sequences that differ greatly from other human DNA, as shown in studies by Lachance et al. (2012), Hammer et al. (2011), and Plagnol and Wall (2006).
In one study, DNA from ancient people in Central Africa (from 8,000 to 3,000 years ago) showed that these individuals had DNA mostly from Central African hunter-gatherers (like the ancestors of the Pygmies) and did not share the archaic DNA found in the Yoruba and Mende. This confirmed that differences in DNA between East, Central, and Southern African hunter-gatherers and West African groups are real. Another study by Lipson et al. (2020) examined DNA from six more ancient African fossils (from the last 18,000 years) and found that their DNA came from Southern, Central, and Eastern African hunter-gatherers, but none had the archaic DNA found in the Yoruba.
A 2020 study suggests that 2% to 19% of DNA in four West African populations may have come from an unknown ancient human group that split from the ancestor of modern humans and Neanderthals between 360,000 and 1.02 million years ago. This study also found that some of this ancient DNA is present in non-African populations, and the mixing of DNA with this ancient group happened between 0 and 124,000 years ago. This includes a time before humans left Africa and before the split between African and non-African populations. Another recent study found genetic differences in African populations that existed before modern humans and were later lost in most non-African populations.
Archaic hominins in Eurasia
Hominins were present in Eurasia at least 2 million years before present. Later, when Neandertals and Denisovans began expanding into Eurasia, the continent was already home to descendants of these early hominins. Their genes mixed with those of Neandertals and Denisovans, and eventually, these genes also entered the DNA of modern humans.
Studies of DNA show two major events where genes from superarchaics mixed with others. This suggests that during the late middle Pleistocene, Eurasia had at least two different groups of ancient hominins living there.
A study by Roger et al. (2020) describes an event where Neandersovans (the common ancestor of Neanderthals and Denisovans) mixed genes with a group of superarchaic hominins. This group had been separated from African hominins for at least 2 million years.
Earlier research found a more recent event of gene mixing. Around 350,000 years ago, the DNA of an "erectus-like" hominid was added to the Denisovan lineage. Since these two groups had been separated for about 2 million years before they interbred, they were more distantly related than any other human populations known to have interbred.
Related studies
In 2019, scientists used artificial intelligence to study genes and found signs of a new type of human ancestor in the genes of people today. This ancestor was not Neanderthal or Denisovan.